Damage Control

Proteins, like this ligase, repair broken DNA.

Image courtesy of Tom Ellenberger, Washington University School of Medicine in St. Louis.

Broken DNA—be it a gap, a nick or a double strand break —is a hallmark of cancerous or aging cells. But once key kinases and their regulators detect DNA damage, a molecular cascade swings into action, triggering posttranscriptional changes in proteins that eventually spur enzymes, such as ligases, to repair the damage and facilitate further replication. Until recently, little was known about exactly how mammalian cells sense DNA damage and coordinate these responses and which molecules act to set things right.

By the mid-1990s, researchers had explored the DNA damage response (DDR) using simple model organisms, such as yeast, and found that DDR was a signal transduction pathway with two kinases, namely ATM (ataxia telangiectasia mutated) and ATR (ATM and Rad3-related), orchestrating a suite of phosphorylation events. About 100 proteins were implicated in the response, and about 25 of those were known to be actual kinase phosphorylation substrates.

Then Stephen Elledge, Steven Gygi and a team of researchers from Harvard Medical School irradiated cells with ionizing particles to trigger DNA damage. The team calculated the rate of phosphorylation in irradiated versus undamaged human embryonic kidney cells by immunoprecipitating phosphorylated peptides with attached phospho-antibodies from two cell cultures: One grown in the presence of isotopic (or heavy) amino acids. Using mass spectrometry to sort out light and heavy peptides, the researchers found that phosphorylations increased more than fourfold after irradiation and uncovered more than 700 proteins that were activated either directly or indirectly by ATM and ATR in response to DNA damage. Their results, as described in this month's Hot Paper, were published in a May 2007 issue of Science. ¹

Turning up more than 700 potential phosphorylation substrates was a great surprise, Elledge says. "We probably expected it to be 15% of that number." The paper was "a huge explosion of data" where "whole pathways that weren't known at all," were identified, says Karlene Cimprich, a molecular biologist at Stanford University. "It was an enormous endeavor that led to a wonderful resource of potential effectors and regulators of the DNA damage response," she says. "It's something we immediately pored over."

Upon publication of the Hot Paper, researchers first set out to verify that the specific proteins turned up by the screen were in fact ATM and ATR phosphorylation substrates. "Through work in a large number of labs, it's now clear that this analysis was extremely accurate," says Cambridge University molecular biologist Steve Jackson.

One of the paper's big surprises, says Massachusetts Institute of Technology biochemist Michael Yaffe, was that proteins, such as splicing factors, involved in RNA posttranscriptional modification seemed to be phosphorylated in response to DNA damage, in addition to proteins involved in DNA replication, recombination, cell cycle management, and gene expression. "I don't think anybody expected those to come up," he says. Researchers in Canada have recently shown that an RNA helicase helps repair damaged genomes by clearing defective RNA at the site of double strand breaks. ²

Now researchers are seeking to describe the conformational and molecular changes that occur in conjunction with and downstream of phosphorylation events following DNA damage. Yaffe's lab, for example, is focusing on three kinases—Chk1, Chk2, and Mk2—that are downstream from ATM and ATR and serve as "checkpoint kinases" to advance the DNA damage response as it ripples through the cell. Yaffe says that his group is using a systems biology approach, including mass spectrometry and RNA interference, to characterize the mechanisms at play in DDR and to map out the intricacies of DDR response networks. ³

Cimprich's lab is currently doing a genome-wide siRNA screen to search for specific genes that may contribute to DNA damage. By overlapping her genetic data with Elledge and Gygi's dataset of protein substrates, she hopes to identify new mediators of DNA damage. Similarly, Yaffe's group is cross-correlating RNAi screens, which indicate a phenotypic response in proteins involved in genetic disruption, with the Harvard group's data to link specific kinases to their substrates and describe the DDR phosphorylation cascade in more detail.

There are hints that fully understanding the DDR process will have clinical relevance. For example, Elledge's screen implicated several proteins in the insulin-IGF-1 signaling pathway, suggesting that DNA damage could play a role in age-related metabolic disorders or in diabetes. A team from South Dakota University recently found that decreased ATM expression may lead to insulin resistance in rats fed a high fat diet. ⁴

"DNA damage seems to go hand-in-hand with premature aging," Cimprich says, and ATM and ATR deficient mice have been shown to prematurely age. Her group continues to pursue individual genes behind some of the proteins that the Hot Paper identified.

"There is the potential that some of these proteins that are phosphorylated could turn out to be drug targets in their own right," says Jackson. In addition, particular proteins that are phosphorylated in response to DNA damage could be used as biomarkers to detect the onset of cancer or other maladies tied to malfunctioning genetic material, Jackson adds. Elledge's own lab, for example, has already altered proteins in the Fanconi anemia pathway, which turned up in his proteomic screen, to treat the disease. ⁵

But before clinical applications appear there is much still to learn about how cells mitigate DNA damage. Phosphorylation events, for example, are only one flavor of posttranscriptional modification, Jackson says. DNA damage also triggers protein ubiquitylation, acetylation and sumoylation. "We might, as yet, be looking at the tip of the iceberg."